Abstract
Chronic obstructive pulmonary disease (COPD) is one of the leading causes of disability and death in the world. Many people with disease are undiagnosed or untreated due to lack of consensus about the definition of COPD. The objective was to estimate the prevalence of COPD in the Danish population aged 45–84 years and to identify the proportion of persons not receiving appropriate medical treatment. A population-based study in the two Danish counties (population aged 45–84 years = 299,000 persons) based on data from 155 general practices in the 2004–2006 period. Stratified sampling was used to select 4,757 subjects from the Danish Civil Registration System and the National Health Service databases. The diagnosis of COPD was defined as FEV1/FVC < 0.70 after bronchodilator. The classification of disease severity was based on criteria developed by the Global Initiative for Chronic Obstructive Lung Diseases (GOLD) criteria. COPD prevalence was associated with sex, age, and smoking status. We found an overall COPD prevalence of 12% (95% CI 11%-13%). Standardized to the Danish population, the true prevalence of COPD is 9% (95% CI 8%-10%). The highest prevalence was observed among current smokers (23%) and former smokers (17%) as well as in the older age group (total = 18%; men = 21%; women = 15%). Most subjects with COPD had mild-to-moderate disease. Among subjects with severe and very severe COPD, 25% were not treated with bronchodilator or inhaled corticosteroids. In conclusion we found that the population prevalence of COPD is high among smokers. A large proportion of persons with COPD did not receive appropriate medical treatment.
INTRODUCTION
Lung diseases characterized by chronic airflow limitations not fully reversible by bronchodilation, i.e., chronic obstructive pulmonary diseases (COPD) cause major morbidity and mortality throughout the world (Citation[1]). The airflow limitation is usually progressive and associated with an abnormal inflammatory response to noxious particles or gases in the lungs. Cigarette smoking and passive smoking constitute the most important risk factors for COPD (Citation[2], Citation[3], Citation[4]). Others include occupational dust and indoor and outdoor air pollution (Citation[5]). In Denmark the number of deaths due to COPD has increased by 50% since 1985. COPD is recognized as the fourth most frequent cause of death and this is likely an underestimate due to underreporting (Citation[6]). Danish women have the highest documented COPD mortality rate in the world, with the number of COPD-related deaths more than doubling since 1985 (Citation[7]). The proportion of current smokers in Denmark is 25%, similar to other European countries.
COPD develops silently over years without clinical symptoms. Post-bronchodilator forced expiratory volume per second (FEV1) is correlated to mortality in COPD which emphasizes the importance of early detection, as this makes intervention possible that can stop disease progression. Screening of smokers aged > 40 in general practice suggests that 10%–20% have undiagnosed COPD (Citation[8]). An accurate diagnosis requires spirometry (Citation[9]). The aims of this study were to estimate the prevalence of COPD in the Danish population aged 45–84 years and to determine the prevalence of cases without appropriate medical treatment.
MATERIALS AND METHODS
Recruitment and subjects
The Danish Civil Registration System was used to select a sample of persons aged 45–84 from the population of 299,000 persons in the target age group in two northern Danish counties (North Jutland and Viborg). The participants consisted of a mixed urban/rural population. The sample was drawn in 2004. The selection form was based on the expected prevalence of COPD in 10-years groups (Citation[10]). We computed a needed sample size of 4,850 subjects to obtain a standard error less than 1.5% of the estimated prevalence and decided a sample size of 13,000 subjects to be large enough to meet the demand. The unique 10-digit civil registration number assigned to Danish citizens at birth allowed direct linkage to different population-based registers in the counties. The persons selected were invited by mail to participate in the North Jutland COPD Prevention Study, with one follow-up reminder sent to non-respondents. The invitation letter suggested them to contact to their general practitioner. The participating general practitioner had in advance received a list of their patients who had been selected for the study. Either the physician or a member of his staff, trained in pulmonary function testing, performed the diagnostic tests for the study. In Denmark approximately 99% of all citizens have access to medical care provided by a general practitioner free of charge. The study was approved by the Regional Scientific Ethics Committee for Viborg and North Jutland Counties and the Danish Data Protection Agency. Written informed consent was obtained from all participants.
Data collection
The participants completed a questionnaire including family history of lung and heart diseases, co-morbidities, smoking habits, alcohol consumption, education, work exposures, and allergies. Spirometry testing was performed before, and when relevant, after bronchodilation. Forced expiratory volume per second (FEV1) and forced vital capacity (FVC) were measured using the general practitioners own spirometers. The participants were instructed to perform the blows according to European Respiratory Society (ERS) and American Thoracic Society (ATS) (three acceptable maneuvres, the best of two of which are reproducible) (Citation[11], Citation[12]). Volume and time calibration was performed before study start and every six months thereafter. The calibration procedure was done by trained staff using a 1 litre syringe. The results of spirometric tests were all seen by one of the authors (JGH) for quality control. The highest FEV1 and FVC measurements were used in the analysis both as absolute values and as a percentage of predicted values based on values from normal subjects published by the Danish Respiratory Society (Citation[13]).
To investigate the proportion of patients with COPD who received appropriate treatment we extracted data from the National Health Service on all prescriptions for oral corticosteroid (ATC group H02A), and for asthma and COPD drugs (ATC group R03) i.e. beta-agonists, anti-cholinergics, leukotriene inhibitors, theophyllins and inhaled corticosteroids in the six months before study entry. The National Health Service partly refunds the patients expenditures on these drugs. All pharmacies in North Jutland and Viborg are equipped with electronic systems that record the personal identification number, the drug, dose, and date each drug is dispensed. This prescription information was transferred to a research database at Aarhus University.
Definition of COPD
Obstruction was defined as a FEV1/FVC ratio < 0.70 after bronchodilation. Severity was defined in terms of the percentage of predicted values based on normal subjects in four categories based on post-bronchodilator values of FEV1 (Citation[1]), mild obstruction: FEV1− > 80% of the predicted value, moderate: 50%–79% of the predicted value, severe: 30%–49% of the predicted value, and very severe: < 30% of the predicted value (Citation[1]). For participants whose FEV1/FVC ratio was < 0.70 of the predicted value a reversibility test was performed using eight inhalations of Combivent (1 dose = 100 μ g salbutamol and 20 μ g ipratropium) followed by spirometry after 30 minutes. A positive bronchodilation response was defined as normalization of the FEV1/FVC ratio > − 0.70 of the predicted value.
Statistical methods
The prevalence of COPD was defined initially as the number of subjects with obstruction, based on pre-bronchodilation values of FEV1 and FVC, divided by the total number of subjects. Then prevalence was calculated more broadly as the number of subjects with obstruction both before and after bronchodilation. Finally, the prevalence was standardized to the Danish population as of January 2006. The 95% confidence intervals (95% CI) were calculated using a normal approximation. The sample was stratified by sex, age in 10-years groups (45–54, 55–64, 65–74, and 75–84 years), and smoking status (never, former, current). COPD was classified according to disease severity (mild, moderate, and severe). While a fourth category, “very severe” was planned initially, small numbers made it necessary to include the few cases in this category in the “severe” group. The association between smoking status, sex and disease severity was tested for trend using the Cochran-Armitage Trend Test and Kruskal-Wallis test. The significance level was set at 5%. Statistical analyses were performed using SAS software (Version 9.1, SAS Institute, Cary, NC).
RESULTS
Descriptive data
The study period was from August 2004 to June 2006 and 155 general practitioners participated. From the decided sample of 13,087 persons 4,757 (36%) entered the study, close to the estimated needed sample size of 4,850 subjects. 23% were current smokers. Median number of pack years was 35 for men, and 27 for women (P <. 0001), 43% were ex-smokers, and 34% had never smoked. shows the demographic characteristics for persons invited to participate and those who were examined at the general practices. Because the study database did not include information on exact age at the time of study invitation, the age of non-respondents was estimated based on their age as of July 1, 2005 (middle of the recruitment period). Less than 1% of the tests did not fulfill the ERS and ATS goals for good quality test session, and were excluded.
Table 1 Baseline data for all subjects invited to participate in the study. Percentage and number (in brackets)
Prevalence
The pre-bronchodilation prevalence of obstruction among the participants was found to be 20% (95% CI 19%-21%). Of these 19% had a positive bronchodilation response and 222 (5%) subjects had no post-bronchodilation values why they were excluded for further analysis. shows the unweighted results of pre- and post-bronchodilation values. Thus, the overall study prevalence of COPD was 12% (95% CI 11%-13%) based on 4535 persons (4757-222). For women 7% (95% CI 6%-9%) and men 15% (95% CI 13%-16%). Standardized to the Danish population, the true prevalence of COPD was 9% (95% CI 8%-10%). For women 7% (95% CI 6%-8%) and men 11% (95% CI 9%-12%)
Table 2 Prevalence of COPD according to prebronchodilator (n = 4.757) and postbronchodilator (n = 4.535) values
and present the severity of lung function impairment in women and men using GOLD categories (Citation[1]). The results reported are unweighted. According to smoking habits, the prevalence of COPD among female never smokers, former smokers and current smokers was 2% (95% CI 1%-4%), 10% (95% CI 7%-13%), and 14% (95% CI 11%-18%), respectively. The corresponding percentages for men were 4% (95% CI 3%-6%), 15% (95% CI 13%-17%), and 25% (95% CI 22%-29%. There was a significant association for COPD severity with smoking, male gender, and older age (total COPD prevalence = 18%; men = 21%; women = 15%).
Figure 1 Severity of lung function impairment in 1.655 women. Classification suggested by the Global Initiative for Chronic Obstructive Lung Disease (GOLD). The results reported are unweighted. NS = Non smokers, ES = Ex-smokers CS = Current smokers.
Figure 2 Severity of lung function impairment in 3.102 men. Classification suggested by the Global Initiative for Chronic Obstructive Lung Disease (GOLD). The results reported are unweighted. NS = Non smokers, ES = Ex-smokers CS = Current smokers.
Treatment
The cumulative proportions of prescriptions for bronchodilators and corticosteroids, among COPD patients in the six months before study inclusion, by severity, were as follows: Mild 14% (10–19), moderate 31% (25–37), severe 63% (52–72) and very severe 58% (36–77). For inhaled corticosteroids (ICS), the corresponding frequencies were 10% (7–15), 29% (23–35), 53% (43–64), and 58% (36–77). For oral corticosteroids, the corresponding frequencies were 6% (4–11), 7% (5–11), 31% (23–42), and 37% (19–59). Among patients who were treated with ICS 14% with mild COPD and 26% with moderate COPD informed that they had had a previous diagnosis of asthma
DISCUSSION
In this population-based study, we found that the overall prevalence of COPD after standardization to the Danish population was 9%. The highest prevalence was as expected observed among current smokers and ex-smokers and in the older age groups. Most cases had mild to moderate disease. The proportion of patients prescribed asthma/COPD medication during the six months prior to study entry was relatively low, indicating significant under diagnosis or under treatment. Surprisingly it is documented that use of ICS to moderate COPD patients is prevalent in this population, even though it has not been proven clinical relevant in this stage of COPD (Citation[1]). As mentioned above a proportion of these patients gave information about that they had had previously asthma.
The validity of our prevalence estimates might be influenced by several factors. Spirometric data were missing for 222 subjects who did not perform a post-bronchodilation test. Excluding these subjects with a positive pre-bronchodilation test might underestimate the true prevalence, since some of the subjects would fulfil the diagnostic criteria. for COPD. Our results might be influenced by selection bias. There was a preponderance of younger and middle-aged cases compared to cases aged 75 or more. This might have lowered the prevalence estimate, as COPD is more common among elderly than younger people (Citation[14]).
We used the GOLD criterion to identify obstruction as an FEV1/FVC ratio < 0.70. Although this is an acceptable pragmatic approach, it may lead to misclassifying some “never smokers” age over 70, as having COPD because the ratio declines with increasing age (Citation[15]). Still, this possible misclassification should not have a major impact on prevalence, as the proportion of “never smokers” over 70 years old with COPD is relatively small.
Our limited data on non-respondents due to decision of the Ethics Committee did not to allow us to interview them and might have introduced bias. Non-participants in surveys tend to have higher rates of smoking and other unhealthy lifestyle factors (Citation[16]), and therefore there might be an underestimation of COPD prevalence. However, it is also possible that persons with COPD symptoms might be more motivated to participate in a pulmonary function survey. The prevalence of current smokers in this study corresponds very well to Danish population (Citation[17]) and the European population as a whole (Citation[18]).
The general practitioners own spirometers were used for pulmonary testing. Although frequently calibrated, misclassification can not be excluded. However, by using a 1.0 L syringe filled with air at ATPS most sources of instrument inaccuracy and faults will be detected. The improvement in hardware and technical control has reduced the technical sources of variation. Low experience and inadequate spriometry training are reported to be the most common causes of error (Citation[19]). In this study the spirometric tests were performed in accordance with the ERS/ATS references. We therefore think that the risk of misclassification in this study is not higher than in other population based studies. The reason why we used the combination of ipratropium and salbutamol (Combivent) is that the combination of the two drugs results in a greater response in the pulmonary function reversibility testing than ipratropium or salbutamol alone (Citation[20]). Furthermore we used a relatively high dose to ensure a maximum of bronchodilation.
Our findings are in accordance with those from a systematic review and meta-analysis of population-based prevalence estimates published during the 1990–2004 period. It produced a pooled prevalence estimate of 7% (95% CI 6%-9%) based on 67 studies. Thirteen of these used GOLD criteria for severity. The pooled estimate for the 26 spirometrically based studies was 9% (95% CI 7%-11%) (Citation[14]). Prevalence estimates were significantly higher in persons aged 40 years or older (9%), smokers (15%), and males (10%). There was significant heterogeneity in the meta-analysis, which was incompletely explained by subgroup analysis broken down by age and smoking status. Only four studies used a sampling strategy allowing calculation of results representative of the population.
The BOLD Study from 12 sites world wide was a population-based sample (N = 9425) of participants aged 40 years or older at 12 different sites (Citation[21]). The prevalence of GOLD stage 2 or more was 10%. There was heterogeneity in the prevalence and staging of COPD both across sites and between men and women. A Swedish study based on a random sample of adults aged 20–69 years reported a 14% overall prevalence of COPD, using GOLD criteria. The prevalence of COPD was significantly related to age > 45 years (17%) and smoking (24%), but no gender difference was observed (Citation[22]). A Korean population-based epidemiologic survey using a representative national sample aged > 18 years also based on the GOLD criteria found an 8% overall prevalence. In adults > 45 years old the prevalence was reported to be 17% (men: 25%; women: 9%). Age > 65 years, male sex, smoking > 20 pack-years, and low income were predictors for COPD (Citation[23]). In the Health Survey for England a nationally representative population sample over the age of 35 years yielded a spirometry-confirmed prevalence of COPD of 13% (95% CI 12%-14%) based on GOLD criteria. Among people with severe or very severe COPD, only 46% (95% CI 39%-54%) reported any diagnosed respiratory disease. 35% (95% CI 32%–37%) with COPD were smokers compared to 22% (95% CI 21%-23%) of those without COPD (Citation[24]). The pre-bronchodilation prevalence of obstruction was found to be 20% which is in accordance with the results from the Copenhagen City Heart Study (Citation[25]).
In the majority of these studies, COPD severity was mild or moderate in approximately half of the participants. This is similar to our findings. Most of the participants reported no respiratory symptoms, which indicates that COPD is under diagnosed (Citation[22], Citation[24]). This may reflect tolerance of respiratory symptoms among the patients or disregard of symptoms by their doctors (Citation[26]).
Clinical implications
Our study has implications for the diagnosis, prevention and treatment of COPD in general practice. Increasing awareness of COPD in the general population and specifically among current smokers would help in the early diagnosis of this disease, and reduce its impact of COPD on health-related quality of life. There is a particular need to identify middle-aged smokers with undiagnosed COPD. As screening is rather difficult to implement in general practice (Citation[27], Citation[2]), case findings are likely to be a more suitable approach in this setting. As soon as patients at risk have been identified, they should be carefully followed with regular spirometry and encouraged to give up smoking. It is well documented that smoking cessation leads to a better prognosis for patients with COPD at any stage of the disease (Citation[4], Citation[23]).
We sincerely thank the participating general practitioners and their staff for collecting the data. We are also grateful to Elisabeth Kristoffersen, secretary for the Department of Clinical Epidemiology, for her gracious help.
REFERENCES
- Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of Chronis Obstructive Pulmonary Disease, http://www.goldcopd.org.2004(updated 2005)
- Rennard S I, Vestbo J. COPD: The dangerous underestimate of 15% (comment). Lancet 2006; 367: 1216–1219
- Burrows B, Knudson R J, Cline M G, Lebowitz M D. Quantitative relationships between cigarette smoking and ventilatory function. Am Rev Respir Dis. 1977; 115: 195–205
- Løkke A, Lange P, Scharling H, Fabricius P, Vestbo J. Developing COPD: a 25 year follow up study of the general population. Thorax 2006; 61: 935–939
- Liu Y, Lee K, Perez-Padilla R, Hudson N L, Mannino D M. Outdoor and indoor air pollution and COPD –related diseases in high- and low-income countries. Int J Tuberc Lung Dis 2008; 12: 115–127
- Jensen H H, Godtfredsen N S, Lange P, Vestbo J. Potential misclassification of causes of death from COPD. Eur RespirJ 2006; 28: 781–785
- Juel K, Døssing M. Koli Danmark. Statens Institut for Folkesundhed. København 2003. COPD in Denmark, National Institute of Public Health, Copenhagen 2003
- Tinkelman D G, Price D B, Nordyke R J, Halbert R J. COPD screening efforts in primary care: what is the yield?. Prim Care RespirJ 2007; 16: 41–48
- Pellegrino R, Viegi G, Brusasco V, Crapo R O, Burgos F, Casaburi R, Coates A, van der Grinten C P, Gustafsson P, Hankinson J, Jensen R, Johnson D C, MacIntyre N, McKay R, Miler M R, Navajas D, Pedersen O F, Wanger J. Interpretative strategies for lung function tests. Eur RespirJ 2005; 26: 948–968
- Lange P, Groth S, Nyboe J, Appleyard M, Mortensen J, Jensen G, Schnoor P. Chronic obstructive lung disease in Copenhagen: cross-sectional epidemiological aspects. J Intern Med 1989; 226: 25–32
- Quanjer P H, Tammeling G J, Cotes J E, Pedersen O F, Peslin R, Yernault J C. Official statement on lung volumes and forced ventilatory flows. Eur RespirJ 1993; 6(Suppl. 16)5–40
- American Thoracic Society. Standardization of spirometry:1994 update. Am J Respir Crit Care Med 1995; 152: 1107–1132
- Danish Respiratory Society. Spirometry—a recommendation (in Danish). Danish Respiratory Society, Copenhagen 1986
- Halbert R J, Natoli J L, Gano A, Badamgarav E, Buist A S, Mannino D M. Global burden of COPD; systematic review and meta-analysis. Eur RespiJ 2006; 28: 523–532
- Hansen J E, Sun X-G, Wasserman K. Spirometric criteria for airway obstruction. Chest 2007; 131: 349–355
- Snow W M, Connett J E, Sharma S, Murray R P. Predictors of attendance and dropout at the Lung Health Study 11 years follow-up. Contemp Clin Trials 2007; 28: 25–32
- Pedersen S T, Christensen H R, Døssing M. Behandling af nikotinafhængighed. (in Danish) Ugeskr Læger 2002; 164: 1480–1484
- Schroeder S A. What to do with a patient who smokes. JAMA 2005; 294: 482–487
- Ferguson G T, Enright P L, Buist A S, Higgins M W. Office spirometry for lung health assessment in adults: a consensus statement from the National Lung Health Education Program. Respir Care 2000; 45: 513–530
- Dorinsky P M, Reisner C, Ferguson G T, Menjoge S S, Serby C W, Witek T J. The Combination of ipratropium and albuterol optimizes pulmonary function reversibility testing in patients with COPD. Chest 1999; 115: 966–971
- Buist A S, McBurnie M A, Vollmer W M, Gillespie S, Burney P, Mannino D M, Menezes A MB, Sullivan S D, Lee T A, Weiss K B, Jensen R L, Marks G B, Gulsvik A, Nizankowska-Mogilnicka E. International variation in the prevalence of COPD (The BOLD Study): a population-based prevalence study. Lancet 2007; 370: 741–750
- Lindberg A, Jonsson A C, Ronmark E, Lundgren R, Larsson L G, Lundback B. Prevalence of obstructive pulmonary disease according to BTS, ERS, Gold and ATS criteria in relation to doctors diagnosis, symptoms, age, gender, and smoking habits. Respiration 2005; 72(5)47–49
- Kim D S, Kim Y S, Jung K-S, Chang J H, Lim C-M, Lee J H, Uh S-T, Shim J J, Lew W J. Prevalence of Chronic Obstructive Pulmonary Disease in Korea. Am J Respir Crit Care Med 2005; 172: 842–847
- Shahab L, Jarvis M J, Britton J, West R. Prevalence, diagnosis and relation to tobacco dependence of chronic obstructive pulmonary disease in a nationally representative population sample. Thorax 2006; 61: 1043–1047
- Løkke A, Fabricius P G, Vestbo J, Marott J L, Lange P. Forekomst af kronisk obstruktiv lungesygdom i København. (in Danish). Ugeskr Læger 2007; 169: 3956–3960
- Renwick D S, Connolly M J. Prevalence of treatment of chronic airways obstruction in adults over the age of 45. Thorax 1996; 61: 164–168
- Schayck C P, van Loozen J MC, Wagena E, Akkermanns R P, Wesseling G J. Detecting patients at a high risk of developing chronic obstructive pulmonary disease in general practice: cross sectional case finding study. Br MedJ 2002; 324: 1370